Comparison of intelligent approaches for cycle time prediction in injection moulding of a medical device product

Kariminejad, Mandana, Tormey, David, Huq, Saif, Morrison, Jim and McAfee, Marion (2021) Comparison of intelligent approaches for cycle time prediction in injection moulding of a medical device product. In: 37th International Manufacturing Conference, 7 - 8 September 2021, Athlon Institute of Technology, Online.

[img]
Preview
Text
Final Manuscript-Prediction_of_Cycle_Time_in_an_Injection_moulded_Part_with_ANN_and_ANFIS.pdf - Accepted Version

Download (663kB) | Preview

Abstract / Description

Injection moulding is an important industry, providing a significant percentage of the demand for plastic products throughout the world. The process consists of many variables which directly or indirectly influence the part quality and cycle time. The first step in optimizing the process parameters is identifying the most significant variables affecting the desired output. For this purpose, various Design of Experiments methods (DOE) have been developed to investigate the effect of the experimental variables on the output and predict the required settings to achieve the optimal value of the output. In this study we investigate the application of DOE for a commercial injection moulded component which suffers from a long cycle time and high shrinkage. The Taguchi method has been used to analyze the effect of four input variables on the two output variables: cycle time and shrinkage. The component has been simulated in the Moldflow software to validate the predicted output and optimized settings of the variables from the DOE. Comparison of the simulation result and the predicted value from the DOE illustrated good accordance. The calculated optimal setting with the Taguchi method reduced the cycle time from 40s to about 23s and met the shrinkage criteria for this commercial part.

Item Type: Conference or Workshop Item (Paper)
Uncontrolled Keywords: injection moulding; cycle time; ANN; ANFIS; MSE
Subjects: 600 Technology > 620 Engineering & allied operations
600 Technology > 670 Manufacturing
Department: School of Computing and Digital Media
Depositing User: Saif Huq
Date Deposited: 04 Oct 2021 10:37
Last Modified: 04 Oct 2021 10:37
URI: http://repository.londonmet.ac.uk/id/eprint/6993

Downloads

Downloads per month over past year



Downloads each year

Actions (login required)

View Item View Item